The design of artificial proteins with tailor-made catalytic activities represents a longstanding goal in the molecular sciences, with potentially transforming implications for chemistry, biology, and medicine. One general strategy for creating protein catalysts is to computationally design an arbitrary catalytic activity into a protein scaffold and then optimize the activity of this artificial enzyme through directed evolution. Existing methods for the directed evolution of proteins are not ideal for the evolution of enzymes because of their inabilities to apply direct selection pressure for intermolecular substrate binding and confer an evolutionary advantage for multiple turnover catalysts. This proposal addresses these limitations by integrating in vitro compartmentalization, bacterial cell libraries, and fluorescence-activated cell sorting (FACS) into a general selection system for the directed evolution of protein catalysts of bond-forming reactions. The first aim of this proposal is to engineer an Escherichia coli bacterium that is capable of secreting an enzyme while simultaneously displaying a synthetic small molecule substrate on its surface. The next aim is to develop a general selection scheme for catalysts of bond-formation, based on the compartmentalization of the bacterial cells that were engineered to secrete a candidate enzyme that can act upon substrate molecules displayed on their cell surfaces. Bond-formation within the compartments results in the attachment of an affinity handle to the cell surface, which can be detected using fluorescent antibody labeling. Bacterial cells exhibiting the highest levels of fluorescence, thereby encoding the most active multiple turnover enzymes, are isolated by FACS for more rounds of selection. To validate the proposed selection scheme, a model selection enriching for bacterial cells that secrete active phosphopantetheinyl transferase will be performed. The last aim of this proposal is to evolve highly efficient, artificial protein catalysts of a Diels-Alder cycloaddition reaction, using the proposed selection scheme to optimize initial designs generated by the laboratory of Professor David Baker. Catalysts will be designed for the model reaction between a 2-pyrone and an _-alkynyl ester, and the catalytic efficiencies, substrate specificities, and regioselectivites of the best catalysts isolated from the selections will be determined. PUBLIC HEALTH RELEVANCE The development of high-efficiency protein catalysts will benefit public health by eventually enabling a more economical, less wasteful, and more environmentally-friendly production of pharmaceuticals and other important chemicals. Enzymes also promise to improve public health as important tools for environmental remediation.